Repository logo
 
Publication

Autonomous temporal pseudo-labeling for fish detection

2022-06-10Journal article Open access
Simple item page
dc.contributor.authorVeiga, Ricardo
dc.contributor.authorExposito Ochoa, Iñigo
dc.contributor.authorBelackova, Adela
dc.contributor.authorBentes, Luis
dc.contributor.authorParente Silva, João
dc.contributor.authorSemiao, J.
dc.contributor.authorRodrigues, João
dc.date.accessioned2022-07-14T10:29:08Z
dc.date.available2022-07-14T10:29:08Z
dc.date.issued2022-06-10
dc.date.updated2022-06-23T12:11:57Z
dc.description.abstractThe first major step in training an object detection model to different classes from the available datasets is the gathering of meaningful and properly annotated data. This recurring task will determine the length of any project, and, more importantly, the quality of the resulting models. This obstacle is amplified when the data available for the new classes are scarce or incompatible, as in the case of fish detection in the open sea. This issue was tackled using a mixed and reversed approach: a network is initiated with a noisy dataset of the same species as our classes (fish), although in different scenarios and conditions (fish from Australian marine fauna), and we gathered the target footage (fish from Portuguese marine fauna; Atlantic Ocean) for the application without annotations. Using the temporal information of the detected objects and augmented techniques during later training, it was possible to generate highly accurate labels from our targeted footage. Furthermore, the data selection method retained the samples of each unique situation, filtering repetitive data, which would bias the training process. The obtained results validate the proposed method of automating the labeling processing, resorting directly to the final application as the source of training data. The presented method achieved a mean average precision of 93.11% on our own data, and 73.61% on unseen data, an increase of 24.65% and 25.53% over the baseline of the noisy dataset, respectively.pt_PT
dc.description.sponsorshipPOSEUR-03-2215-FC000046
dc.description.versioninfo:eu-repo/semantics/publishedVersionpt_PT
dc.identifier.citationApplied Sciences 12 (12): 5910 (2022)pt_PT
dc.identifier.doi10.3390/app12125910pt_PT
dc.identifier.eissn2076-3417
dc.identifier.urihttp://hdl.handle.net/10400.1/17993
dc.language.isoengpt_PT
dc.peerreviewedyespt_PT
dc.publisherMDPIpt_PT
dc.relationAlgarve Centre for Marine Sciences
dc.relationAlgarve Centre for Marine Sciences
dc.relationLaboratory of Robotics and Engineering Systems
dc.relationCentre for Marine and Environmental Research
dc.relationTurning Marine Protected Areas more efficient by adjusting their zonation and fishing practices
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/pt_PT
dc.subjectEnvironmental monitoringpt_PT
dc.subjectMarine fishespt_PT
dc.subjectObject detectionpt_PT
dc.subjectFish detectionpt_PT
dc.subjectPseudo-labelingpt_PT
dc.subjectUnderwater videopt_PT
dc.subjectDeep learningpt_PT
dc.titleAutonomous temporal pseudo-labeling for fish detectionpt_PT
dc.typejournal article
dspace.entity.typePublication
oaire.awardTitleAlgarve Centre for Marine Sciences
oaire.awardTitleAlgarve Centre for Marine Sciences
oaire.awardTitleLaboratory of Robotics and Engineering Systems
oaire.awardTitleCentre for Marine and Environmental Research
oaire.awardTitleTurning Marine Protected Areas more efficient by adjusting their zonation and fishing practices
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F04326%2F2020/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDP%2F04326%2F2020/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F50009%2F2020/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/LA%2FP%2F0101%2F2020/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT//UI%2FBD%2F151307%2F2021/PT
oaire.citation.issue2pt_PT
oaire.citation.startPage5910pt_PT
oaire.citation.titleApplied Sciencespt_PT
oaire.citation.volume12pt_PT
oaire.fundingStream6817 - DCRRNI ID
oaire.fundingStream6817 - DCRRNI ID
oaire.fundingStream6817 - DCRRNI ID
oaire.fundingStream6817 - DCRRNI ID
person.familyNameVeiga
person.familyNameExposito Ochoa
person.familyNameBelackova
person.familyNameBentes
person.familyNameParente Silva
person.familyNameSemião
person.familyNameRodrigues
person.givenNameRicardo
person.givenNameIñigo
person.givenNameAdela
person.givenNameLuis
person.givenNameJoão
person.givenNameJorge
person.givenNameJoao
person.identifier1603578
person.identifier2266994
person.identifierR-001-F67
person.identifier.ciencia-idD212-85A6-C85A
person.identifier.ciencia-id821B-029B-B7DA
person.identifier.ciencia-idC619-63A6-6076
person.identifier.ciencia-idB616-9C41-C169
person.identifier.ciencia-id8A19-98F7-9914
person.identifier.orcid0000-0002-7557-8304
person.identifier.orcid0000-0003-1921-006X
person.identifier.orcid0000-0002-9718-4250
person.identifier.orcid0000-0001-6884-2886
person.identifier.orcid0000-0001-7046-9646
person.identifier.orcid0000-0002-7667-7910
person.identifier.orcid0000-0002-3562-6025
person.identifier.ridD-5057-2009
person.identifier.ridL-6700-2015
person.identifier.scopus-author-id57203130604
person.identifier.scopus-author-id6603195176
person.identifier.scopus-author-id15924042200
person.identifier.scopus-author-id55807461600
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
rcaap.rightsopenAccesspt_PT
rcaap.typearticlept_PT
relation.isAuthorOfPublication60cb9fb6-94b2-4657-935e-3e7eea696a49
relation.isAuthorOfPublicationaece6782-b285-456f-acec-81e5bcf40592
relation.isAuthorOfPublication9262c5e2-ef1e-4475-989b-e5359561d37b
relation.isAuthorOfPublication6569b202-4643-45ec-9861-9fd8d5f09601
relation.isAuthorOfPublication64e5e52a-7d80-46c2-8f83-fb159685c195
relation.isAuthorOfPublication12454e89-e25e-4e96-8b64-39974f6fba07
relation.isAuthorOfPublication683ba85b-459c-4789-a4ff-a4e2a904b295
relation.isAuthorOfPublication.latestForDiscoveryaece6782-b285-456f-acec-81e5bcf40592
relation.isProjectOfPublicationfafa76a6-2cd2-4a6d-a3c9-772f34d3b91f
relation.isProjectOfPublication15f91d45-e070-47d8-b6b8-efd4de31d9a8
relation.isProjectOfPublication63f1f0ee-a2d4-4055-8a65-111048e05495
relation.isProjectOfPublication794d4c77-c731-471e-bc96-5a41dcd3d872
relation.isProjectOfPublication07379456-b4eb-4283-9f9f-e78ccb10bd15
relation.isProjectOfPublication.latestForDiscovery794d4c77-c731-471e-bc96-5a41dcd3d872

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
applsci-12-05910.pdf
Size:
38.15 MB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
3.46 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

CCM2-Artigos (em revistas ou actas indexadas)